Document Details

Document Type : Thesis 
Document Title :
STEGANALYSIS ALGORITHM FOR PNG IMAGES BASED ON FUZZY LOGIC TECHNIQUE
خوارزمية اكتشاف المعلومات المخفية في الصور PNG بالاعتماد على تقنية المنطق الضبابي
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : Professional criminals need the ability to exchange messages confidentially, and as a result, have exploited the rapid advances in information and communication technology. A prevalent method of doing so is Steganography – the process of hiding a secret message into media. The message can be embedded into any medium (text, image, audio or video). To detect hidden information, tools are used for discovery and analysis. As a counter-measure, tools have been developed in order to detect hidden information form digital media such as text, image, audio or video files. Images (PNG, JPEG, GIF, and BMP) are famously used for steganography. Research in the field has revealed that there are few pre-existing studies done on PNG images and this research will contribute to the body of knowledge by undertaking an increased focus on the PNG format. An experiment was conducted which showed that there are narrow gaps hindering the ability of stenographic tools to detect hidden elements. As such, this research aims to design an algorithm based on artificial intelligence (AI) that is able to detect hidden information embedded by any steganography tool in PNG images. However, the efficiency and performance of previous approaches found in the fields literature have shown room for improvement. In this research, we focus on algorithm design for optimum efficiency of hidden message detection in PNG files. In more detail, the techniques examined are a novel hybrid model developed based on Adaptive Neuro-Fuzzy Inference Systems of the Sugeno Type (ANFIS) and Multi-Layer Perceptrons (MLPs) techniques, Support Vector Machines (SVMs), Neural Networks (Multi-Layer Perceptrons MLPs) and Adaptive Neuro-Fuzzy Inference Systems of the Sugeno Type (ANFIS). These techniques are compared on the basis of the resulting confusion matrices, as well as by using the Receiver Operating Characteristic (ROC) curves. Finally, we introduce our message detection system for PNG files based on the Least Significant 
Supervisor : Prof. Dr. Daniyal Mohammed Alghazzawi 
Thesis Type : Master Thesis 
Publishing Year : 1439 AH
2018 AD 
Added Date : Thursday, March 8, 2018 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
جواهر عبدالله القحطانيAlqahtani, Jawaher AbdullahResearcherMaster 

Files

File NameTypeDescription
 43162.pdf pdf 

Back To Researches Page